CAST Proposal and Status Report 139th Meeting of the CERN SPSC
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
CAST Proposal and Status Report 139th Meeting of the CERN SPSC K. Zioutas, University of Patras, on behalf of CAST Collaboration and external collaborators M. Vogelsberger (MIT) & A. Kryemadhi (Messiah U.) (CERN-SPSC-2020-022 / SPSC-SR-277 02/10/2020 http://cds.cern.ch/record/2738387 ) CERN, 13th October 2020 1
ABSTRACT With less than 2 months of data taking, the CAST-CAPP cavity experiment managed to cover a significant fraction of previously unexplored parameter DM axion space in the range 19.7 to 22.4 μeV (4.7 to 5.5 GHz), and exceeded the previous limits by ADMX-Sidecar by far. CAPP is a sub-detector of CAST with 4 resonance cavities. It is the full realization of the conceptual idea from 2011 and transforms CAST from an axion helioscope to an axion haloscope. Since 2019 data are taken with individual cavities as well as with all 4 cavities phase matched (=coherently). The experiment is not yet running under optimal conditions and further optimizations are already being prepared. These will improve the sensitivity even further and will lead to world-leading results. The full previously granted data taking time of 6 months, which due to the COVID19 pandemic would extend into 2021, is essential to achieve the planned sensitivity. To fully exploit its sensitivity, we have recently installed a second spectrum analyser which measures simultaneously the ambient electromagnetic noise. CAST-CAPP has pioneered the new technology of fast scanning and phase-matching of multiple cavities, which is capable to detect transient events from streaming dark matter or axion mini-clusters. An unambiguous detection of such events requires an independent detector at another location. We are currently in contact with the IBS Korea on the possibilities to facilitate establishment of a second detector there. From the CAST report to SPSC: http://cds.cern.ch/record/2738387/files/SPSC-SR-277.pdf 3
Solar axions: NATURE Phys.13(2017)584 CAST Motivation: Dark Universe …. Axions strong CP problem & DM & solar axions …. Chameleons DE Direct DM search @ CERN CAST’s approach: 1 out of 3 methods 4
CAST P. Sikivie g Working principle: • a-helioscope Axion - source E2 × • a-haloscope ~100 ktons of axions/s overlooked? ↓ Axion - conversion Eaxion 5
Initially: Solar axions (DM) solar chameleons (DE) 2013- < 4 keV > < 1 keV> 2003- K.Z., P. Brax PR D85(2012)043014 DM axions (
CAST – CAPP >> DM axions + transients Cavity axion-haloscope https://arxiv.org/abs/1703.01436 https://doi.org/10.1111/j.1365-2966.2011.18224.x https://doi.org/10.1088/0004-637X/814/2/122 https://doi.org/10.1093/mnras/stx1474 Theses: Marios Maroudas Kaan Özbozduman 10
FAST TUNING & 4 PHASE-MATCHED CAVITIES! Increase sensitivity by combining coherently the read-outs from several cavities: = ∙ So far:: 0 h TOTAL Phase Matched cav’s: 448.9 h (18.7days) & 125 MHz 11
BACKGROUND DATA (B=OFF) Goal: Comparison of frequencies with significant power excesses between B=ON & OFF So far:378.3h (15.8d) • ies: 4.2h (0.2days ) TOTAL BKG: 442.9 h (17.6 days) & ~150MHz POWER Frequency => In time domain 12
RESULTS FROM ALL DATA 13
RESULTS Histogram of the grand spectrum with a gaussian fit (in red). The outliers being scrutinized are the ones on the right side of the red curve. outliers 14
BACKGROUND (ENVIRONMENTAL EMC NOISE) POWER 07/2020 Frequency => Therefore: A simultaneous independent measurement with a second spectrum analyzer connected to an external antenna measuring the EM noise in the CAST area = A MUST: NOW INSTALLED! frequency 15
& ENVIRONMENTAL EMC NOISE Simultaneous comparison of data from cavities with EMC data Both are in the frequency domain: FFT by the SA Cavity 4 EMC noise in CAST area Power excess The noise bump Frequency => Frequency => 16
So far data-taking time with single & PM cavities and EXCLUSION PLOT ASSUMING GALACTIC DM AXIONS B=ON: 1461 hours (60.9 days). Spikes correspond to longer measurement times. For comparison ADMX is also given progress 09/2019 10/2020 17
FAST TUNING (SINGLE CAVITIES) Fast cavity tuning takes advantage of possible DM transients: streams / clusters gravitational (self-)focusing flux enhancement. Special case free fall with low speed: ~46 km/s at 1 AU! [Adrien Leleu] In total with single cavity and B=ON: 1013.2h (42.2 days) & ~660 MHz Tidal streams 18
SENSITIVITY PLOT FOR DM AXIONS INCLUDING FLUX ENHANCEMENT Due to gravitational lensing (by sun, …., Moon) or axion mini-clusters Improvement due to streaming DM axions or axion-miniClusters lasting ~1 hour and for an increased flux by factor 103±1 parasitically sensitive also to cosmological axion transients 10/2020 19
Comparison with ADMX around 0.5 - 0.8 GHz CAST (660 MHz) vs. ADMX (200 MHz) @~5 GHz @ volume effect 20
DATA TAKING • Projected relic axion detection sensitivity with the remaining 4 months (of the previously granted 6 months in total) of data taking. • Modest Flux enhancement due to streams / clusters reaches sensitivity to QCD axions. 21
CAST – CAPP conclusions 1. ~2 months of data taken over the summer. We managed to cover a significant fraction of previously unexplored parameter axion space in the range of 19.7 to 22.4 μeV and exceeded the previous limits by ADMX-Sidecar by far. 2. Data taken with individual cavities as well as with all 4 cavities phase matched (=coherently). 3. More optimizations to extend the sensitivity even further are already being prepared and will result to world-leading results. 4. The full previously granted data taking time of 6 months, which due to COVID-19 would extend into 2021, is essential to achieve the planned sensitivity. 5. To fully exploit its sensitivity, we have recently installed a second spectrum analyser which measures simultaneously the ambient EM noise. 6. CAST-CAPP has pioneered the new technology of fast scanning and phase-matching of multiple cavities, which is capable to detect transient events from streaming dark matter or axion miniclusters. 7. an unambiguous detection of such events requires an independent detector at another location, therefore we are currently in contact with IBS Korea on the possibilities to facilitate the establishment of a 2nd detector there. 22
Other CAST activities in parallel and parasitically to CAST-CAPP 23
CAST – Rades >> DM axions >>> R&D for babyIAXO. Cavity axion haloscope Theses: Sergio Arguedas Cuendis Jessica Golm José María García Barceló 24
RADES 1m long cavity For 2020 data taking • Increase of the detection volume by a factor 5 using a more complex cavity geometry relying on alternating irises theoretically shown to be more scalable than the previous cavity. The damaged cryogenic amplifier was replaced and has been functioning as expected. 25
RADES 1m long cavity measurement Axion peak simulation • The 2020 cavity Q-value is lower than expected. • Its EM response is also more complex and requires further characterization which are currently taking place. • Data taking with this cavity is ongoing. • Important R&D for babyIAXO. 26
CAST-RADES preliminary exclusion limit CAST-RADES exclusion limit using 2018 data-set in the context of other haloscope experimental results. Our result provides the strongest exclusion limit for an axion mass above 30 µeV. 27
CAST Micromegas detector Solar axion search(DM) with a new detector & the 2nd XRTelescope: first xenon data Important R&D for any future solar axion search , e.g. babyIAXO. Theses: Cristina Margalejo Blasco, Héctor Mirallas 28
Xenon runs - calibration Fe55 spectrum: Hit channels Fe55 spectrum: All channels 3 keV Xe + 50% Ne + Energy threshold Resolution @ Gain (5.9 keV 2.3% isobutane 5.9keV peak in ADC) Hit channels 1100 eV 37% 5000 NATURE Phys.13(2017)584 All channels 700 eV 19% 8000 • Small peak at low energies (1 - 1.5keV). Possible fluorescence? • Simulations will help define its energy. • It will be useful for a better energy calibration at low energies. • The argon fluorescence peak at ~3 keV no longer appears, reducing background level in that energy range. >> IMPORTANT! 29
Xenon runs – background spectrum 224.5 h Preliminary Includes only 1-prong events Raw spectrum (113.5h) Generic cuts Generic cuts and veto 3 keV 3 keV 6x10-3 c/keV/cm2/s 2.1x10-6 c/keV/cm2/s 1.2x10-6 c/keV/cm2/s Background level for 1-prong events after generic cuts based on 55Fe calibrations with Xenon NATURE Phys.13(2017)584 30
Summary • Ar+2.3% isobutane 2019 data taking campaign • ~120 hours of tracking • Background ~2x10-6 c/keV/cm2/s • Analysis tools under development to achieve higher efficiency • Ar+2.3% isobutane 2020 data taking campaign • ~35 hours of tracking • Improved electronics configuration • Xe+50%Ne+2.3% isobutane 2020 data taking campaign • 35 hours & 23 trackings • Saving all channels allows us to have higher gain, better energy resolution, lower energy threshold, and potentially better X-ray selection algorithms Next steps • Develop new analysis strategy that matches the new electronics configuration • Understand the differences we observe among different data taking campaigns and with the X-ray tube • Perform simulations for the different phenomena we observed • Possible re-calibration of the detector in the X-ray tube 31
InGrid Integrated Grid solar chameleons (dark energy) & low energy solar axions (DM) with a sensitive low energy x-ray detector 32
InGrid Data analysis of the 2017/18 data nearing completion on the software side. Analysis framework including limit calculation is done. Background rate still preliminary, due to: • usage of detector vetoes via scintillators to be finalized (cut values to be fixed) • GridPix(= Timepix + InGrid). Outer 6 GridPix validation for background rate over whole chip outstanding (=> understand clusters near chip edges - data lossy, gap between chips) • Possible improvement from better FADC utilization • significant variation of photo peak position (in charge or pixels) of ⁵⁵Fe calibration data observed. Current energy calibration suffers possible bias because of this. Excess of 2017/18 in plot near 4 and 9.5 keV likely due to overestimate of energy in part of dataset. Finally, a structured check of the whole analysis pipeline is ongoing to verify no major bugs are hiding, by validating all intermediate steps. 33
Status of the KWISP detector Working group: J. Baier, D. Božičević, G. Cantatore, S. Cetin, H. Fischer, A. Gardikiotis, M. Karuza, Y. Semertzidis, K. Zioutas Detection by radiation pressure on a Si3N4 nano-membrane Theses: Justin Baier Martin Markanovic Dorotea Bozicevic 34
KWISP current version overview KWISP 3.5 • Fabry-Pérot interferometer with full fiber-optic beam transport • improved monolithic optics design - Already acquired ~ 9h • uncoated membrane of sun-tracking data • “sandwich-type” passive - Data taking continues vibration isolation • built-in force calibration with auxiliary beam • DMD chopper KWISP 1.5 • Michelson interferometer - Short sun-tracking run in January 2020 • monolithic design • high density Pt coated - Presently under test membrane and calibration in • “sandwich-type” passive the laboratory vibration isolation • “chopper-less” detection 3 KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 35 5
KWISP 1.5 status • Parasitic tracking performed in January 2020 with KWISP 1.5 • Sensitive to a static force ⇒ change in the frequency of the membrane mechanical resonant mode • Now under calibration in the laboratory to check for long term pressure and temperature effects • Preliminary force sensitivity in the pN range FORCE => TIME => Sample KWISP 1.5 data from the 23/1/2020 sun tracking run The linear trend can be interpreted as due to the variation of the component of the gravity force normal to the membrane surface when the CAST magnets tilts vertically 3 KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 36 6
KWISP 3.5 calibration • KWISP 3.5 is home made • stable lock under vacuum • successful direct force calibration with built in auxiliary 532 nm laser beam (amplitude-modulated at a controlled frequency) • Preliminary sensitivity estimate from direct calibration 8∙10−12 N⁄ √Hz @ 2.5 kHz “no” peak applied force peak External force OFF External force ON Direct force calibration near the chopper expected frequency of operation (~2.5 kHz) KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 37
KWISP 3.5 preliminary sun-tracking runs • July 2020: Installation • August 2020: Sun-tracking- parasitically, no magnetic field needed • Sun tracking data: 11 hours. • Background & Calibration: 21.3 hours. • Preliminary sensitivity 28 pN/√Hz • No solar signal yet • Minimum force detected in 275 s was ~3.3 pN, MORE DATA AVAILABLE: ~30000 sec. Therefore, • >10× better min detected force calibration run 5σ level tracking run 5σ level Avg. noise level Avg. noise level calibration frequency chopper frequency KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 38
KWISP next steps >>> parasitically • KWISP 3.5 • Complete analysis of taken data • faster online calibration (under way) • continue taking solar data as long as possible • KWISP 1.5 • study long term stability KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 39
CAST conclusions • ~2 months of data taken over the summer, now stopped up to mid of November due to water tower maintenance (not related to CAST). • Fast tuned & phase matched cavities pioneered by CAST. • The full previously granted data taking time of 6 months, which due to COVID-19 would extend into 2021, is essential to achieve the planned sensitivity. • The missing 4 months of data taking are extremely important for the theses of our students. The estimated remaining data taking time for 2020 is less than 1 month. For the DM search the efficiency is ~18h/24. 40
Backup slides KWISP - 139th SPSC meeting - CERN, Oct. 13th 2020 41
BEYOND THE ~5KHZ SIGNAL SEARCH Histogram of the combined spectrum with a gaussian fit (in red). Combined spectra are more appropriate for DM transients due to non-restriction to a ~5kHz signal search. 42
RESULTS Combined spectra are more appropriate for transients (no width constraint). 43
You can also read